JPS60125053A - Data communication system - Google Patents

Abstract

PURPOSE:To evade an unnecessary fallback to a lower electrical transmission rate, shorten the time required for a procedure, and shorten a communication time by dividing a specific signal into plural blocks, and testing a transmission speed with the number of errors of every block. CONSTITUTION:A code 1 is control signal and a DSC signal in G3 mode, and pieces of information on the electrical transmission rate and possible fallback mode of a transmission side are contained in a facsimile information field. A training signal 2 is for equalization, convergence, etc., and then a training check (TCF) signal 3 consisting of successive 0s follows them. When a reception side succeeds in training with said three signals, an an acknowledge signal 4 is transmitted. This is a reception setup confirmation signal in G3 mode. When the reception side fails in the training, an NOT signal such as a train failure signal is transmitted. A training check is made while the TCF signal 3 is divided into plural blocks to find data errors, block by block.

Description

DETAILED DESCRIPTION OF THE INVENTION [Technical Field] The present invention relates to a data communication system, and particularly to a data communication system that tests a transmission rate according to the number of errors in a received predetermined signal.

[Prior Art] In conventional data communication systems, such as G3 facsimile machines that comply with the CCITT (Consultative Committee for International Telegraph and Telephone) recommendations, multiple transmission rates are selected in descending order according to the state of the transmission path in the pre-transmission procedure. So-called fallback modes are known to test and determine a suitable transmission rate.

Generally, in this test, so-called training, a training signal for converging the automatic equalizer of the modem on the receiving side and a TCF (training check) signal are concatenated and transmitted. This TCP signal is data “0”
is a modulated signal that continues for 1.5±0·15 seconds. Conventionally, the method involves receiving a TCP signal and checking that there are no data errors in an interval longer than a certain time (for example, 1 second), or that the number of erroneous data in the entire training check interval is less than a certain value. A training check was conducted.

However, in general, the equalizer of the modem on the receiving side converges over time, so in the above method, even if it converges in the latter half of the TCP signal, there is a high possibility that it will be judged as a train failure, and the lower The drawback was that it would fall back to the transmission rate of .

[Purpose] The present invention has been made in view of the above problems, and prevents unnecessary fallback to a lower transmission rate.
The purpose is to provide an excellent data communication method that allows communication at faster transmission speeds.

[Example] Hereinafter, the present invention will be described in detail based on the example shown in the drawings. However, in the following, a G3 facsimile machine conforming to the CCITT standard will be described as an example. In the following explanation, C
For known signals in the CITT recommendations, only the signal names are written and detailed explanations are omitted.

FIGS. 1 and 2 show the training according to the present invention. FIG. 1 shows a case where the training was successful, and FIG. 2 shows a case where the training failed.

In both figures, what is indicated by the symbol l is a control signal, specifically the DO3 signal in G3 mode. As is well known, the PIF (Facsimile Information Field) of this control signal includes information such as the transmission rate and the possible fall pack mode of the transmitter.

The training signal 2 performs known equalizer convergence, and is followed by the TCF signal 3, which is a series of "0"s. The above control signal 1, training signal 2, and TCF signal 3 are sent from the image transmitting side to the receiving side.

If the receiver succeeds in the training consisting of the above signals, an affirmative signal 4 is transmitted as shown in FIG. G
In the 3 mode, this signal is specifically a CFR (Confirm Ready to Receive) signal. If the receiving side fails in the training, a negative signal 5 is transmitted as shown in FIG.

Specifically, this signal is the FTT (lane failure) signal.

In the present invention, the TCP signal 3 is divided into N blocks BKI to BKn as shown in FIG. 3, and a training check is performed by examining data errors in each block. The number of bits of l block is determined as appropriate, but
Here, it is assumed that there are P pieces.

As mentioned above, TCP signal 3 is a continuous transmission of "0", but if the line condition makes it difficult for the receiving modem's equalizer to equalize it and the convergence is not completed, rOJ will be generated even if it is demodulated.
cannot be identified. Assuming that one block of blocks BK+ to BKn is P bits, the number of errors in each block BK, to BKn, that is, the number of data that is not "0", is ERl and ER2.
,...ERn, the data error rates in each block BK, ~BKn are ER1/P and ER2/P, respectively.
,...ERri''/P.

It is thought that errors occur in a pattern like the one shown in FIG. 4 in the TCP signal section. In FIG. 4, the error rates ERI/P, ER2/P, . . . ERn/P of each of the above blocks are plotted according to possible patterns.

In pattern 10 of FIG. 4, the error is more than half over the entire interval of the TCP signal, and it is clear that the equalizer has not converged.

On the other hand, in the case of pattern 7, the equalizer converges in the first half of the TCP signal section. Further, in the case of patterns 8 and 9, convergence is completed in the latter half of the TCP signal, and in the cases of these patterns 7 to 9, good results should be obtained even if the transition to image communication continues. However, in the conventional system, when the number of turns is 8.9, the number of errors is checked in the entire TCP interval, so it is determined that the train has failed, and the transmission rate falls back to a lower transmission rate.

Therefore, in the present invention, an error rate of 7 is determined for each block as shown by diagonal lines in FIG. do.

With this configuration, patterns 8 to 9 in FIG.
Even in this case, there is no need to fall back to a lower transmission rate, so the transmission speed can be increased, the procedure time can be shortened, and the overall communication time can be shortened.

Next, a specific configuration for performing the above determination operation will be explained.

FIG. 6 is a block diagram showing the configuration of a facsimile machine for both sending and receiving purposes. In FIG. 6, reference numeral 21 indicates an image reading section using a CCD sensor or the like. 22
is a recording section, which is composed of a thermal printer or the like.

The data read or recorded by the reading section 21 and the recording section 22 described above is processed by a known method by a control section 2o constituted by a microcomputer or the like. The read image data undergoes processing such as encoding in the control unit 20, is sent to the transmission unit 23, is modulated, and is sent to the network control unit 25.
The signal is sent out to a transmission path such as the telephone line 26 via the telephone line 26. Further, data transmitted from the transmission path is input to the receiving section 24 via the network control section 25, demodulated, guided to the control section, decoded, and recorded in the recording section 22.

Control of pre-transmission procedures and post-transmission procedures is performed by the control section 2o via the transmitting section 23 and the receiving section 24.

Generally, the transmitting section 23 and the receiving section 24 are integrally configured as a modem, and the receiving section 24 includes the above-mentioned automatic equalizer.

The calculation of the error rate associated with the reception of the TCP signal and the success or failure of training are performed by the control section 20 according to the procedure shown in the flowchart of FIG.

When the control device 20 detects the TCP signal transmitted via the network control section 25 and the reception section 24, the control device 20 shifts to the control from step 31 onward in FIG. In step S1 in FIG. 7, "1" is stored in a counter configured by software or hardware in a part of the memory included in the control device 20. This counter is used for each block BK of the TCP signal 3 mentioned above.
, ~BKn.

In step S2, the control device loads the received n-th block data into a predetermined area of the memory, calculates the number of errors ERn, that is, the number of data that is not "0", and divides this value by the number of blocks P. error rate ERn
/P is calculated.

In step S3, this error rate ERn/P is compared with the maximum allowable error rate Kn of the block, and ERn'/P≦K
In the case of n, the process moves to step s4, and ERn/P>Kn
In this case, the process moves to step S7. In this judgment operation, if the maximum error rate that differs for each block is made into a table and stored in a predetermined area of the memory in advance as shown in FIG. Kn can be determined by reference.

In step S4, the block number counter is incremented by 1, and then in step S5 it is determined whether the TCP signal has ended. If the TCP signal has not ended, the process returns to step S2 and moves to the determination operation for the next block of the TCP signal. If the TCP signal has ended, the process moves to step S6, and the affirmative signal 4, ie, the CFR signal, is sent out as shown in FIG.

If the error rate exceeds a predetermined number in step S3, the end of the TCP signal is waited for in step S7, and then, in step S8, a negative signal 5 indicating train failure, that is, an FTT signal, is sent as shown in FIG. Send out.

By dividing the TCP signal into blocks as described above and checking for errors in each block, if the equalizer of the receiving section 24 converges in the TCP signal section, it will not be considered as a train failure, and it will be unnecessary. The present invention can be implemented with a small amount of software changes as described above, and is therefore very effective. It's easy and cheap.

Although the G3 facsimile machine has been described above as an example, the present invention can of course be applied to communication systems that perform similar training.

In addition, although the TCP signal is divided into N blocks of P bits in the above, it is not necessary to divide each block into blocks of equal number of bits, and blocks of unequal bits are considered appropriate for determination by those skilled in the art. You can divide it into Also, in the control shown in Figure 7, the error rate is calculated for each block, but if each block has the same number of bits, success or failure can be determined based on the number of errors without calculating the error rate each time. good.

Furthermore, although the criteria for the error rate shown in FIG. 5 is defined by a linear function, it is not limited to this, and any criterion defined by a curve that is considered appropriate by those skilled in the art may be determined.

[Effects] As is clear from the above description, according to the present invention, in a data communication system that tests the transmission speed according to the number of errors in a received predetermined signal, the predetermined signal is divided into a plurality of blocks. Since it uses a configuration that tests the transmission speed based on the number of errors in each block, the simple and inexpensive configuration avoids unnecessary fallbacks to lower transmission rates, shortens the time required for the procedure, and reduces communication time. It is possible to provide an excellent data communication method that can shorten the time and reduce running costs.

[Brief explanation of the drawing]

1 and 2 are explanatory diagrams showing the state of communication in the present invention, respectively, and FIG. 3 is a TCP diagram in FIGS. 1 and 2.
An explanatory diagram showing the state of signal division, Fig. 4 is a diagram showing various error occurrence patterns in the TCP signal section, Fig. 5 is a diagram showing the error rate criteria in the present invention, and Fig. 6 is a diagram showing the present invention. FIG. 7 is a block diagram showing the configuration of a facsimile apparatus employing the above, and FIG. 7 is a flowchart showing the control procedure of the control section in FIG. 2... Training signal 3... TCF signal 4... Affirmation signal 5... Negative signal 20... Control section 22... Recording section 23... Transmission section 24... Receiving section 25. ...Network control unit Fig. 1 Fig. 2 Fig. 3 U-P, I- Fig. 4n Fig. 5

Claims (1)

[Claims] A data communication system that tests a transmission speed according to the number of errors in a received predetermined signal, characterized in that the predetermined signal is divided into a plurality of blocks and the transmission speed is tested according to the number of errors in each block. Communication method.